My involvement with the BLAST programs this year was minimal. I consulted with the BLAST team from the IEB several times on ways to modify the BLAST algorithm so that increasing the number of protein alignment results requested does not result, in some cases, in completely new alignments appearing at the top of the results list. This problem arises because BLAST re-scores alignments in order to deal with non-standard amino acid compositions, and must retain more more the number of results requested because some may move up in rank after re-scoring. The number of results retained was not always sufficient to avoid the reported problem, and I proposed a new method for calculating how many results to retain. No publication resulted from this work.